Search Results (1,394)

Background: Antimicrobial-resistant Salmonella enterica from chicken sources is an important One Health concern because poultry can serve as a route for transmission of resistant foodborne pathogens. This study analyzed whole-genome sequencing data from 4611 chicken-associated Salmonella enterica isolates in the United States from the National Center for Biotechnology Information (NCBI) Pathogen Isolates Browser across seven antimicrobials: tetracycline, streptomycin, sulfisoxazole, ampicillin, ceftriaxone, nalidixic acid, and chloramphenicol. Methods: Principal component analysis, hierarchical clustering, and frequency-based bar-plot analysis were used to characterize gene-occurrence patterns and prioritize antimicrobial resistance (AMR), virulence, and stress-response genes. These prioritized genes were evaluated by logistic regression, and genotype–phenotype discrepancy analysis examined resistant isolates lacking expected antimicrobial-specific AMR genes using Fisher’s exact tests with Benjamini–Hochberg false discovery rate correction. Results: Resistant isolates shared broad genomic features: mdsA and mdsB were the most frequently detected AMR genes across all seven groups; virulence genes including invA, iroB, iroC, sinH, pipA, and avrA were repeatedly prominent; and golS, golT, and asr formed a common stress-response background. In the logistic regression models that predicted the probability that each Salmonella enterica isolate was resistant to a specific antimicrobial based on the presence or absence of selected genes, AMR-only models showed area under the receiver operating characteristic curve (AUC) values of 0.926–0.995, combined virulence and stress-response models showed AUC values of 0.826–0.990, and models combining all three gene categories showed AUC values of 0.935–0.998. Discrepant isolates were uncommon (0.06–2.10%), but enriched virulence and stress-response genes included recurrent pco, sil, spv, and steB signals across multiple antimicrobial groups. Conclusions: These findings identify resistance-associated genomic backgrounds involving virulence and stress-response genes and demonstrate their predictive value when incorporated into gene-based models. The enriched genes provide candidates for future genomic and experimental investigations to test whether they contribute functionally to antimicrobial resistance, persistence, or co-selection in poultry-associated Salmonella enterica. Full article

Fog–cloud systems allow distributed and latency-sensitive IoT applications to share edge, fog, and cloud resources for efficient computation, storing and delivering services. Nonetheless, the current methods of task scheduling and resource allocation tend to be based on deterministic prediction, heuristic schedules, or response optimization, restricting risk sensitivity and responsiveness to dynamic workloads. To address these issues, we propose a graph network with a reinforcement learning framework to avoid congestion and schedule tasks with a low makespan in fog–IoT systems. The work starts with real-time monitoring of queue states, delay, bandwidth, energy and deadline properties of upcoming IoT tasks. Quantile Regression Long Short-Term Memory (QR-LSTM) predicts a normal task flow and congestion based on the task queue. The congestion tasks are further scheduled using a Delay-Aware Influence Reinforced-Graph Neural Network (DAIR-GNN). Then, Topology-Sensitive Resource Influence Propagation is used to map and analyze the relationship between the task sender and receiver within a network. After that, Dual-Stage Risk-Aware Hierarchical Policy (DRHP) combines Proximal Policy Optimization (PPO) to select the kinds of sources, like fog or cloud, based on the topology. Similarly, Model-Agnostic Meta-Learning (MAML) with Soft Actor–Critic allocates the resources of each tasks within the selected server. Both RL models are trained using Few-Shot Adaptive Policy Transfer to make better predictions. Lastly, Confidence-Uncertainty Regulated Exploration (CURE) is used to compute the prediction score for task allocation improvement. The proposed framework achieves a variance ratio of 80.6%, latency is 0.626 s, and the success rate is 98% in the prediction of congestion, scheduling and allocation of resources. These results demonstrate the improved ability to ensure reliable and effective fog–cloud allocation in response to dynamically changing workloads. Full article

Accurate molecular property prediction underpins progress in drug discovery, medicinal chemistry, and molecular design and depends critically on the quality of the learned molecular representations. While Graph Neural Networks (GNNs) and Graph Transformers have advanced this field considerably, most existing approaches emphasize a single structural perspective and struggle to jointly capture local connectivity, higher-order substructures, explicit topology, and long-range dependencies in a single learning process. This work introduces Topology-Guided Multi-Level Graph Learning (TMGL), a unified framework that integrates topology-enhanced node encoding, motif-guided representation learning, a Multi-Scale Graph Isomorphism Network, topology-aware graph transformer attention, adaptive multi-scale feature fusion, and cross-level contrastive learning into a single end-to-end architecture. By coupling topology preservation and motif consistency objectives with hierarchical contrastive alignment across node-, motif-, and graph-level embeddings, TMGL enables these complementary structural views to interact directly rather than being combined post hoc. The framework was evaluated on seven MoleculeNet benchmark datasets spanning binary classification, multi-label classification, and regression tasks. TMGL achieved Receiver Operating Characteristic Area Under the Curve (ROC-AUC) scores of 0.937, 0.960, 0.962, and 0.723 on BACE, BBBP, ClinTox, and SIDER, respectively, and RMSE values of 0.611, 1.293, and 0.436 on ESOL, FreeSolv, and Lipophilicity, outperforming a broad set of state-of-the-art baselines on most classification benchmarks and on Lipophilicity, while remaining competitive on ESOL and FreeSolv. Ablation experiments confirmed that each architectural component contributes meaningfully to overall performance, and embedding visualizations further illustrated the discriminative, class-consistent structure of the learned representation space. These results demonstrate the value of jointly modeling complementary structural perspectives for molecular graph learning and highlight the potential of the TMGL framework for applications in computational chemistry and drug discovery. Full article

Generative artificial intelligence (GenAI) can support educational work while changing how users engage in independent generation and evaluation. This study examines whether the GenAI task use frequency has different associations with perceived role-specific capability and AI dependence. A cross-sectional survey of 1105 respondents (659 students and 446 teachers) from five Beijing universities measured all focal variables by self-report. Hierarchical regressions with HC3 standard errors were complemented by formal inverted U tests, bootstrap intervals, measurement-model sensitivity analyses, and nested out-of-fold prediction. Both capability outcomes showed positive lower-range and negative upper-range slopes, with the maxima located within the observed use range and supported by observations on both sides within each literacy stratum. Between the 25th and 75th percentiles of AI literacy, the model-implied maximum shifted by 0.715 use scale units for students (95% bootstrap CI [0.525, 0.973]) and 0.643 units for teachers (95% CI [0.450, 0.883]). The primary literacy measure was a 12-item mean; four-dimensional and second-order CFA specifications supported its interpretation as a broad composite. Separate exploratory dimension models yielded positive literacy-by-use interactions, with the largest point estimate for critical evaluation. Dependence was positively associated with use, habit, and trust and negatively associated with literacy. The combined squared use and interaction increments for dependence were 0.0024 and 0.0004, with 90% interval upper bounds of 0.0058 and 0.0031, below an exploratory 0.01 $R^2$ benchmark. Belief-centred and trust omission specifications supported the broader reliance pattern, although trust and dependence remained conceptually close. Nonlinear ridge recovered most of the capability prediction improvement over linear ridge. The findings describe between-person associations, not individual optimal use levels or intervention effects. They motivate the evaluation of literacy-oriented task design and complementary approaches to trust calibration and habitual reliance. Full article

Objective: This study aimed to compare social functioning and social communication characteristics between children with CP and CVI and children with CP without CVI, and to identify demographic, clinical, and functional factors independently associated with social functioning. Methods: This observational cross-sectional study included 70 children with CP aged 48–84 months, including 35 children with CP and CVI and 35 children with CP without CVI. Gross motor function was assessed using the Gross Motor Function Classification System (GMFCS), manual ability using the Manual Ability Classification System (MACS), speech performance using the Viking Speech Scale (VSS), social communication using the Social Communication Questionnaire (SCQ), and social functioning using the Social Function domain of the Pediatric Evaluation of Disability Inventory (PEDI). Between-group comparisons were performed, and hierarchical multiple linear regression analysis was conducted to identify factors independently associated with social functioning. Results: Significant between-group differences were found in the distributions of GMFCS (p = 0.007), MACS (p = 0.043), and VSS (p = 0.027) levels. Children with CP and CVI had lower PEDI Social Function scores and higher SCQ total scores than children with CP without CVI (p < 0.001). Across all SCQ domains, including reciprocal social interaction, communication, and restricted, repetitive, and stereotyped behaviors, children with CP and CVI showed higher scores (all p < 0.001). In the hierarchical regression analysis, the final model explained 96.7% of the variance in social functioning (R² = 0.967; adjusted R² = 0.963; p < 0.001). Age (β = 0.089; p = 0.008), CVI status (β = −0.295; p < 0.001), GMFCS level (β = −0.270; p < 0.001), VSS level (β = −0.211; p < 0.001), and SCQ total score (β = −0.272; p < 0.001) were independently associated with social functioning. Conclusions: Children with CP and CVI had lower social functioning and higher scores for SCQ-assessed social communication and behavioral characteristics than children with CP without CVI. The independent associations of CVI status, gross motor function, speech performance, and SCQ total score with social functioning highlight the importance of a multidimensional approach to evaluating social functioning in children with CP. Full article

Background: Regular physical activity is generally associated with favorable body composition; however, little is known about the relationship between long-term folk dance participation, overall physical activity, sedentary behavior, and body composition. This study aimed to determine whether dance training experience, total physical activity, and sitting time were independently associated with BMI and body composition parameters among young adult Polish folk dancers. Methods: This cross-sectional study included 79 members of a Polish folk dance ensemble (51 women and 28 men) aged 18–30 years. Body composition was assessed using bioelectrical impedance analysis. Physical activity and weekday sitting time were assessed using the Polish short form of the International Physical Activity Questionnaire (IPAQ-SF), while dance experience was determined using a study-specific questionnaire. Spearman’s rank correlations and hierarchical linear regression models including sex and age as covariates were used to examine associations between dance experience, physical activity, sitting time, BMI, and body composition parameters. Results: Median dance experience was 30 months, while median total physical activity was 4746 MET-min/week. Dance experience was not significantly correlated with BMI or any body composition parameter. Total physical activity showed a weak positive correlation with BMI (r_s = 0.25, p = 0.025) and basal metabolic rate (r_s = 0.23, p = 0.039) but was not significantly correlated with body fat percentage, fat-free mass, muscle mass, total body water, or bone mass. In hierarchical regression models, dance experience did not significantly improve the explanation of BMI or any body composition parameter. The addition of physical activity and sitting time significantly improved the model for basal metabolic rate (ΔR² = 0.028, p = 0.021), but not for the other outcomes. In the final models, sex was the strongest predictor of body composition, while total physical activity was independently associated with basal metabolic rate (β = 0.17, p = 0.007). Sitting time was not independently associated with any analyzed outcome. Conclusions: Among young, physically active Polish folk dancers, cumulative dance training experience was not associated with BMI or estimated body composition parameters. Sex accounted for a substantial proportion of variability in body composition, whereas total physical activity showed an independent association only with basal metabolic rate. Longitudinal studies incorporating dance-specific training volume, dietary intake, and sleep are warranted. Full article

Background/Objectives: Adenoma detection rate (ADR) is a central colonoscopy quality indicator, but the extent to which annual procedural volume explains physician-level ADR remains uncertain. We evaluated the association of procedural volume with ADR while separating within-physician annual changes from between-physician differences and explored rank-order persistence of physician-level ADR. Methods: We retrospectively analyzed 36,864 institutional ADR-eligible colonoscopies performed at a Japanese hospital from 2013 to 2024. The canonical dataset comprised 171 physician-year observations from 62 physicians. Weighted physician-year regressions decomposed log annual volume into within-physician and between-physician components. Case-level generalized estimating equations, grouped-binomial models, denominator-restricted analyses, influence analyses, a Bayesian hierarchical logistic model, and an exploratory restricted cubic spline analysis were used as complementary analyses. Results: The examination-weighted ADR was 52.7%, and median annual volume was 188 examinations. In the primary model, the within-physician coefficient was −0.003 (95% confidence interval [CI], −0.039 to 0.033; p = 0.882), corresponding to an estimated ADR difference of −0.19 percentage points per doubling of annual volume (95% CI, −2.69 to +2.31). The between-physician coefficient was 0.027 (95% CI, −0.012 to 0.067; p = 0.177), corresponding to +1.90 percentage points per doubling of usual volume (95% CI, −0.86 to +4.66). Grouped-binomial and denominator-restricted analyses did not identify a positive within-physician association. An exploratory pooled spline suggested curvature at lower volumes, but the overall spline association was not statistically significant. Conclusions: No positive within-physician association between annual procedural volume and ADR was identified. The between-physician estimate was imprecise and did not exclude a potentially meaningful positive association. Annual procedure counts should not be used alone as a surrogate for detection quality; direct longitudinal monitoring of ADR with attention to denominator size, uncertainty, case mix, and modifiable examination processes remains necessary. Full article

Background/Objectives: Gastric cancer severely affects health-related quality of life (HRQoL). This study evaluated HRQoL profiles and identified clinical, functional, and symptom correlates among gastric cancer patients in Kazakhstan. Methods: A cross-sectional study was conducted in a single-center cohort (n = 387) using validated Russian-language European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30) (Kazakh verbal clarification provided when needed, n = 12). Hierarchical multivariable linear regression was performed to identify factors independently associated with Global Health Status/QoL (GHS). Results: Mean GHS score was 51.9 ± 24.3. Major reported symptom burdens included stomach area pain, fatigue, and anxiety. In the final adjusted regression model (Step 3), GHS was significantly predicted by older age (ß = −0.21, p < 0.001), pain (ß = −0.37, p < 0.001), body image (ß = −0.12, p = 0.035), and functional domains: physical (ß = −0.20, p = 0.022), role (ß = 0.14, p = 0.026), emotional (ß = 0.32, p < 0.001), and cognitive functioning (ß = −0.14, p = 0.018). Employment categories also remained significant predictors: white-collar (ß = 0.09, p = 0.044), service sector (ß = 0.20, p < 0.001), blue-collar (ß = 0.19, p < 0.001), and retired (ß = 0.16, p = 0.022). Anxiety showed a positive association with GHS (ß = 0.24, p < 0.001), reflecting a statistical suppression effect, whereas fatigue (ß = −0.02, p = 0.745) was not significant after controlling for covariates. Conclusions: HRQoL in gastric cancer patients is strongly associated with physical symptom burden, age, and functional status. Targeted pain management, psychological support, and holistic care should be prioritized in routine oncology practice. Full article

Background/Objectives: Patients with lung cancer often experience impaired quality of life because of symptom burden and psychosocial difficulties. Cancer stigma and distress are associated with poorer quality of life, whereas adaptive coping may support psychological adjustment. This study examined the relationships among cancer stigma, distress, cancer coping, and quality of life and identified factors associated with quality of life in patients with lung cancer. Methods: This cross-sectional correlational study included 168 patients with lung cancer receiving inpatient or outpatient care at a university hospital in the Republic of Korea. Data were collected from 1 August to 1 September 2025, using structured self-report questionnaires. Descriptive statistics, group comparisons, Pearson correlations, and hierarchical multiple regression were performed. Results: The mean scores for cancer stigma, distress, cancer coping, and quality of life were 46.43 ± 8.46, 4.49 ± 2.53, 56.43 ± 15.07, and 74.02 ± 14.93, respectively; 57.1% of participants reported clinically significant distress. Quality of life was negatively correlated with cancer stigma (r = −0.63, p < 0.001) and distress (r = −0.76, p < 0.001) and positively correlated with cancer coping (r = 0.71, p < 0.001). Cancer coping, current activity level, and distress were independently associated with quality of life. The final model explained 75.2% of the variance, with cancer coping showing the largest standardized coefficient. Conclusions: Better cancer coping, higher activity levels, and lower distress were associated with better quality of life. Psychosocial nursing care should support adaptive coping, functional activity, and distress management. Full article

Emerging adulthood is marked by increasing independence and responsibility across educational, occupational, and social domains. During wartime, disruptions to daily routines, security, and perceived control may threaten mental well-being, understood as positive emotional, social, and psychological functioning. These challenges may be particularly salient for individuals with specific learning disabilities (SLD) and/or attention deficit hyperactivity disorder (ADHD), who may experience difficulties in executive functioning and maintaining everyday routines. This study examined differences in mental well-being between emerging adults with and without SLD and/or ADHD and assessed the contributions of executive functioning, health-promoting daily activities, psychological capital, and perceived social support during wartime. Participants were 317 Israeli emerging adults aged 18–28 years. Among the 316 participants with valid diagnostic-group classification, 115 reported SLD and/or ADHD and 201 reported neither condition. Emerging adults with SLD and/or ADHD reported lower mental well-being, greater executive functioning difficulties, and lower engagement in health-promoting daily activities than those without SLD and/or ADHD. Mental well-being was positively associated with psychological capital, health-promoting daily activities, and perceived social support, and negatively associated with executive functioning difficulties. In the hierarchical regression, diagnostic-group status was initially associated with mental well-being but was no longer significant after psychological capital, health-promoting daily activities and executive functioning were entered; perceived social support made a small additional contribution. The final model explained 37.2% of the variance in mental well-being. These findings highlight the relevance of cognitive-functional, psychological, occupational, and social factors to mental well-being during a prolonged national crisis. Full article

Background: Childbirth-related posttraumatic stress symptoms (CB-PTSS) have been associated with impaired mother–infant bonding, but the factors underlying this relationship remain unclear. This study examined whether maternal stress and perceived infant regulatory difficulties (PIRD) independently accounted for the association between CB-PTSS and mother–infant bonding difficulties in a statistical mediation model. Methods: This cross-sectional study included 542 Portuguese mothers of infants aged up to 24 months. Participants completed validated measures of CB-PTSS, maternal stress, perceived infant regulatory difficulties, and mother–infant bonding. Hierarchical regression and parallel mediation analyses were conducted. Results: Higher CB-PTSS were associated with greater maternal stress, more perceived infant regulatory difficulties, and greater bonding difficulties. Perceived infant regulatory difficulties explained additional variance beyond CB-PTSS and maternal stress. Both maternal stress and infant regulatory difficulties showed significant independent indirect effects, while the direct association between CB-PTSS and bonding difficulties remained significant, indicating partial mediation. Given the cross-sectional design, these findings should be interpreted as statistical associations rather than evidence of temporal or causal relationships. Longitudinal studies are needed to establish temporal ordering and directionality. Conclusions: Maternal stress and perceived infant regulatory difficulties may represent key statistical correlates linking traumatic childbirth to bonding difficulties and constitute promising targets for early screening and intervention. Full article

Extreme rainfall in semi-arid mountain regions can occur within broader hydroclimatic states involving persistent precipitation, atmospheric moisture, and terrestrial wetness, yet their respective associations and prospective information remain poorly resolved in data-sparse environments. Here, we investigated the hydroclimatic organization of extreme regional rainfall in the semi-arid Peruvian Andes using satellite-derived and reanalysis observations spanning 2001–2025. Extreme rainfall was defined using the global 99th percentile of three-day accumulated precipitation, yielding 92 threshold-exceedance days that were declustered into 41 independent episodes. These episodes occurred preferentially under wet regional conditions, with the strongest separation from true non-extreme background days observed for multiday precipitation accumulation, followed by atmospheric and soil-moisture anomalies. Thirty-seven of the 41 independent episodes (90.2%) occurred when both total column water vapor (TCWV) and shallow soil-moisture anomalies were positive, corresponding to an odds ratio of 19.86 (year-cluster bootstrap 95% CI: 10.24–83.44). However, adjusted nested models provided no robust evidence for an independent shallow-soil-moisture contribution or a TCWV × soil-moisture interaction, indicating that the compound-wet state is better interpreted as a marker of organized regional wetness than as evidence of atmosphere–land synergy. Under a strictly non-overlapping rolling-origin architecture, hydroclimatic information available before the prediction date provided modest prospective discrimination of future three-day rainfall extremes. The complete hierarchical logistic model achieved a pooled PR-AUC of 0.126 compared with 0.094 for seasonality alone, whereas Random Forest and XGBoost showed no robust improvement over logistic regression. Native-resolution MODIS NDVI analyses likewise provided no robust evidence of post-event vegetation greening, although positive median responses occurred after aggregated ecological rainfall exposures. Overall, independent extreme regional rainfall episodes in the semi-arid Peruvian Andes were strongly organized within wet atmospheric and terrestrial hydroclimatic states, while evidence for atmosphere–soil interaction, strong predictive skill, and subsequent ecological response was substantially weaker. These findings demonstrate the value of integrated Earth-observation records for distinguishing hydroclimatic-state association from prospective prediction and ecological response in data-sparse mountain environments. Full article

Background: Physical activity, screen time, and cardiorespiratory fitness have been associated with cognitive functioning in children; however, their mutually adjusted associations with specific dimensions of selective attention remain insufficiently understood. This study examined the associations of habitual physical activity, screen time, and cardiorespiratory fitness with selective attention in primary school children. Methods: This cross-sectional study included 386 children aged 9–12 years (54.4% girls). Habitual physical activity was assessed using the Physical Activity Questionnaire for Children (PAQ-C), screen time using the Youth Leisure-Time Sedentary Behavior Questionnaire (YLSBQ), cardiorespiratory fitness using the 20 m shuttle run test, and selective attention using the d2-R Test of Attention. Selective attention was characterized by processing speed (TR), percentage of errors (E%), and concentration performance (CON). Associations were examined using correlation and multiple linear regression analyses adjusted for age and sex, followed by hierarchical regression models examining the mutually adjusted associations of physical activity, screen time, and cardiorespiratory fitness with selective-attention outcomes. Results: Higher physical activity was significantly associated with a greater number of total items processed (TR; B = 69.87, β = 0.660, p < 0.001), a lower percentage of errors (E%; B = −1.75, β = −0.356, p < 0.001), and better concentration performance (CON; B = 26.44, β = 0.659, p < 0.001). Greater screen time was primarily associated with a higher error percentage (β = 0.597, p < 0.001), while associations with processing speed and concentration performance were weaker. Higher cardiorespiratory fitness was associated with greater processing speed (β = 0.468), fewer errors (β = −0.455), and better concentration performance (β = 0.667; all p < 0.001). In the final hierarchical models, after mutual adjustment for physical activity, screen time, cardiorespiratory fitness, age, and sex, physical activity was associated with processing speed (β = 0.600, p < 0.001), whereas screen time and cardiorespiratory fitness were not associated with this outcome. For attentional errors, higher screen time (β = 0.523, p < 0.001), lower physical activity (β = −0.117, p = 0.014), and lower cardiorespiratory fitness (β = −0.270, p < 0.001) were associated with a higher error percentage. For concentration performance, both physical activity (β = 0.395, p < 0.001) and cardiorespiratory fitness (β = 0.412, p < 0.001) were positively associated, whereas screen time was not associated with this outcome. Conclusions: Physical activity, screen time, and cardiorespiratory fitness showed distinct associations with specific components of selective attention. Higher physical activity was particularly associated with faster processing speed, higher screen time with greater attentional errors, and both physical activity and cardiorespiratory fitness with better concentration. Given the cross-sectional design, these findings should be interpreted as associations rather than causal relationships. Full article

While the linear and homogeneous effects of the built environment on travel behavior have been widely examined, the nonlinear patterns and contextual heterogeneity remain insufficiently explored—particularly in Chinese megacity suburbs. Recent methodological advances, such as double machine learning and gradient boosting decision trees, have begun to reveal threshold effects and non-linear moderating mechanisms that conventional linear models cannot capture. This study examines the associations between neighborhood-scale built environment characteristics and residents’ travel-related carbon emissions, based on 108 communities in Beijing’s Fengtai District, China. Using a “5D+3G” indicator system and combining hierarchical regression, subgroup regression, and random forest, the analysis identifies exploratory nonlinear patterns in the built environment–emissions relationship. The random forest model showed notable overfitting given the modest sample size (training R² = 0.730; test R² = 0.223; cross-validation mean = −0.172), and the identified thresholds are therefore treated as hypothesis-generating rather than confirmatory. Linear models reveal that most built-environment attributes exhibit weak and statistically insignificant associations with emissions, indicating that average-effect specifications may obscure the complexity of these relationships. Rail-transit accessibility shows distance-based attenuation: emissions remain low within the exploratory range of approximately 500 m of metro stations, increase between 500 and 1500 m, and plateau beyond 1500 m in this sample. POI functional density exhibits diminishing marginal returns, with a sample-specific saturation pattern near 4 POIs/ha. The built environment–emissions relationship thus appears more heterogeneous and context-dependent than previously assumed. Based on these empirical patterns, the study proposes a five-category neighborhood typology and corresponding design intervention strategies tailored to local conditions. The analytical framework offers a transferable reference for architects, urban designers, and planners seeking evidence-based guidance for low-carbon neighborhood design in similar contexts, while the specific thresholds require validation in other settings. Full article

Drawing on the Positive Development and Assessment Competencies (PDAC) model, this cross-sectional study examined whether character virtues and self-reported arts engagement make complementary contributions to primary school students’ self-perceived competencies. Data were collected from 993 fifth- and sixth-grade students; 695 provided valid responses to both arts items. Continuous competency scores were analysed using hierarchical linear regression with HC3 robust standard errors, demographic adjustment, and Benjamini–Hochberg correction. Binary high-score outcomes were retained as sensitivity analyses. After PREFARTI and demographics were controlled, the six-virtue block added 17.2–36.9 percentage points of explained variance across the eight competencies (all q < 0.001). After virtues were controlled, PREFARTI added unique variance in linguistic communication (ΔR² = 0.040, q < 0.001), digital competency (ΔR² = 0.011, q = 0.044), and cultural and artistic competency (ΔR² = 0.065, q < 0.001). Its multivariate contribution to the eight-competency profile was significant, with Pillai’s trace = 0.163, F(24, 1722) = 4.13, p < 0.001. Of the selected interaction sets, only the block for personal autonomy added corrected explanatory value (ΔR² = 0.022, q = 0.023), with heterogeneity concentrated in PREFARTI-by-wisdom (q = 0.009). The joint pattern remained significant after school fixed effects and exclusion of the smallest PREFARTI group, although the univariate digital result was less robust. The findings indicate an asymmetrical pattern of association: character virtues show broad associations with perceived competencies, whereas self-reported arts interest and participation contribute more selectively. These findings do not establish transfer, prediction, or causality. Full article